Implications of Using Lead-Free Solders
on X-ray Inspection of Flip Chips and BGAs
David Bernard, Dage Precision Industries Inc.
Fremont, CA 94538 e-mail: d.bernard@dage-group.com
Abstract
All contract manufacturers (CM) of printed wiring boards
(PWBs) are having to consider their approach to using leadfree solders to meet current, and future, environmental,
legislative and market demands. As these lead-free solders
have markedly different operating parameters from those of
the tried and tested lead-containing solders, the opportunity
for production problems rises. Therefore, the application of
appropriate test and inspection procedures now becomes
even more important to ensuring product quality and
reliability. However, the test and inspection equipment, and
methods, used for decades may not now be appropriate for
this new challenge. This may be because, at worst, some test
parameters are no longer valid with the lead-free solders, or,
less seriously, the inspection control settings must be varied
to accept revised test responses based on acceptable levels
for these new materials. Implementing lead-free solder into
production, therefore, raises questions as to the acceptability
and adequacy of a company’s existing test methods. This
paper discusses the implications of using x-ray inspection on
lead-free versions of flip chips and BGAs.
Introduction
Driven by the concerns for the environmental impact of
increasing amounts of lead containing products within waste
landfill sites, there is now legislative and commercial
pressure on OEMs to implement lead-free solutions within
the printed circuit boards (PCBs) of their products. As
contract manufacturers (CMs) charged with supplying these
OEMs, it is therefore becoming evermore necessary to
consider implementing a lead-free strategy within
production and to decide what effects this change from
traditional lead solders will have on the manufacturing and
test and inspection procedures that are used.
Japan is already well advanced in converting their PCB
manufacturing to using lead-free alternatives and expects to
have completed the switch by the end of 2004. In Europe,
the European Union has set the date of July 2006 for
compliance by all member states, which includes the
newest, mainly former Eastern Bloc members, as well as the
existing 15 countries. In the US, in contrast, no formal date
for a transfer to lead-free production has been raised.
Therefore, whilst supply to the home market may not
require any switch for the forseeable future, the danger of
not implementing lead-free into a CMs manufacturing
capability could result in their being excluded from
supplying into the ‘lead-free’ markets. It should be noted,
however, that although there is a legislative timetable for
Europe to be lead-free, there are still many European CMs
that have yet to implement this capability. Those CMs that
have already changed have been through the extended
learning curve that the new process profiles for these leadfree solders require. As such, there is much information
available to help new lead-free adopters avoid some of the
pitfalls experienced so far. Further information on the
European lead-free roadmap together with other useful data
based on experience of manufacturers moving to lead-free
can be found in the web sites in references 1 and 2.
Traditional solders have been of the Tin/Lead (Sn/PB)
variety and many decades of experience has been invested
into using these materials to the best advantage. This has
defined the process windows within which to produce PCBs
and also shaped the inspection and testing regimes that have
been implemented to ensure quality manufacture. In
particular, the reflow temperature profile used within
surface mount technology production has been defined
around the, typically, 183°C liquidus temperature of these
solders. In addition, the visual look of the Sn/Pb solder after
reflow has been used by experienced operators to confirm
the quality of their solder joints. When using lead-free
solders, these fundamental indices are completely changed
and therefore the underlying tenets that have been
established for so long in PCB manufacture may no longer
be valid.
Lead-free Solders
The most commonly used Sn/Pb solders have consisted of
around 63% Sn and 37% Pb. The most commonly used leadfree alternatives are the, so called, SAC alloys which are
named after the mixture of tin (Sn), silver (Ag) and copper
(Cu) that they contain. Although different solders from
different suppliers have varying elemental ratios, the SAC
alloys all contain around 97% tin. There are other lead-free
solders that are available, with other elemental
compositions, but, once again, their main constituent is Sn.
So, in general, the lead is being replaced by tin in these new
solders. The immediate effect of this change is to
dramatically increase the liquidus temperature of these
solders to around 215 – 230°C. As a result, the reflow oven
temperatures typically used for SMT production must be
raised to a much higher level – around 290°C so as to ensure
the solder reflow will occur. Such higher temperatures then
bring into question the viability of the components used and
bare boards themselves. For example, to be lead-free, a
standard tin/lead HASL finish would no longer be permitted.
Although there is a lead-free HASL finish that comprises of
a tin/copper/nickel combination. A silver board finish is
often suggested for lead-free because of its bonding
properties with the copper pads. Organic solderable
protector (OSP) and gold board finishes may also be
considered for lead-free applications. There is also the
question of the susceptibility of the components to this
additional temperature as well as the finish on the
component legs (reference 2). Finally, the look of these
solders after reflow is not the same as that seen in Tin/Lead
meaning the ‘experience’ gained in this area is no longer
helpful in determining problems. In other words, using leadfree is not just simply a question of changing to a higher
temperature profile within the reflow oven; board type,
finish and component choice all have to be considered. The
implication of using lead-free solders, therefore, is that a
new ‘learning-curve’ has to be climbed with necessary
experimentation and testing to validate these new processes.
However, is the test and inspection equipment typically used
for process control still valid for these new materials? It has
already been mentioned that the look and behavior of leadfree solders is different to that of Tin/Lead. Therefore, with
optical inspection techniques, questions must be asked as to
the efficacy of their light sources, software algorithms, etc.,
for inspecting lead-free materials. In a similar vein, will
existing and new x-ray systems be able to cope? With the
dramatic increase in the use of BGA and CSP components
this question becomes more ever more important as only xrays can inspect where optical systems cannot. To ensure
that x-ray inspection remains valid for lead-free applications
and understanding of the technique is required.
X-ray Inspection
Two dimensional (2-D) x-ray inspection systems, system
most commonly used within the PCB industry, are basically
x-ray shadow micrographs (see figure 1). The technique
demands that an x-ray light source (called an x-ray tube)
produces x-rays, within which the sample is bathed. The
differing materials within the sample absorb more, or less,
of the x-ray radiation depending on their density and atomic
number and cast a shadow of that material at the detector.
The denser the material, then the darker the shadow. In this
way, solder and copper tracks appear dark compared with
the laminated circuit board in a PCB, for example. The
detector converts the incident x-rays into optical images for
the operator to view. The closer the sample is moved to the
x-ray tube then the larger the shadow becomes and this is
how magnification is achieved.
Figure 1: Basic 2-D x-ray system configuration
The quality of the x-ray radiation produced by the x-ray
tube, and its effectiveness as discriminating the different
materials within a sample to produce a useful analytical
image, are defined by the tube settings used. These settings
are called the x-ray tube accelerating voltage, or kV, and the
tube power and are used to set the tube at appropriate levels
so as to get a good contrast image at the detector.
The accelerating potential is the applied voltage between the
anode and cathode of the x-ray tube that makes electrons,
produced at the cathode, strike the target anode, creating xrays. The tube power is calculated from the product of the
accelerating potential and the filament current within the
cathode used to produce the electrons. The more power then
the brighter the x-ray source. However, there are technical
limitations on the maximum power that the tubes can
achieve (see references 3 and 4).
The kV is also a measure of the penetrating power of the xrays. The higher the kV used then the more penetrating are
the resultant x-rays. This means that higher kVs need to be
used to image dense, or thick but relatively less dense,
objects. At lower kVs, only thin and less dense samples can
be inspected. Otherwise, the x-rays have insufficient
penetrating power to travel through the sample and strike the
detector, so creating the image.
The interaction between penetrating radiation and matter,
such as x-rays passing through a sample under test, is not a
simple relationship and is based on a number of factors.
Further detailed information on this matter can be found in
reference 5. However, the important consideration, as far as
imaging lead-free solders, is that absorption of the radiation
increases with atomic number and density of the material
that is present. Table 1 shows the differences in the atomic
number (Z) and the density between the different elements
contained within lead and lead-free solders. Lead has a very
high Z value and this is being replaced in the lead-free
materials mainly with additional quantities of tin that has a
much lower Z value as well as a lower density. Therefore,
the effect of imaging lead-free solders using the same tube
parameters that would currently be used for traditional lead
solders may result in the image being overexposed. This is
because the lower Z / less dense materials in the lead-free
material will absorb the x-rays less, allowing more to pass
through to the detector and potentially saturate it.
Element
Lead
Tin
Silver
Copper
Symbol
Pb
Sn
Ag
Cu
Atomic
Number (Z)
82
50
47
29
Atomic
Mass (amu)
207.2
118.71
107.8682
63.546
maintained during PCB manufacture. As can be seen in the
x-ray images below (images 1 – 8) of lead-free examples of
BGA and CSP investigations, there is sufficient contrast and
resolution to be able to clearly identify the faults that may
occur. Many of these x-ray images come from the results of
the SMART Group Lead-Free Hands on Experience that
was held during Nepcon UK 2003 and were produced on
Dage x-ray inspection systems. At this event, the same PCB
with four different finishes had components placed and
reflowed, either by vapour phase or convection, using a
SAC (tin./silver/copper) lead-free solder. Further
information and results of using these different
combinations can be found in reference 6.
Conclusion
No. of
Protons
82
50
47
29
No. of
Neutrons
125
69
61
35
11.34
7.31
10.5
8.96
Density
(g/cm3)
Table 1: Comparison of fundamental properties of
elements used in lead and lead-free solders
To compensate for this effect, it may be necessary to
decrease the tube kV and/or the power values that have been
used previously for x-ray inspection of the lead version.
This will ensure that suitably contrasted images are still
available for making analysis. However, these changes, if
indeed are needed at all, are likely to be small in quantity.
Values of around 5 – 15 kV less accelerating voltage than
for the equivalent lead soldered board may be the typical
adjustment necessary. The tube power might need to be
decreased by 0.25W but this will be dependent on the x-ray
system being used because of the variation in tube
efficiency, capability and brightness from manufacturer to
manufacturer. Whilst it may be necessary to adjust the x-ray
tube parameters for lead-free application, most modern xray systems have the ability to make comprehensive contrast
adjustments to the captured x-ray images – see reference3.
Therefore, the changes in contrast that using lead-free
solders might cause compared to previous lead versions may
be able to be adjusted electronically without the need to
change the tube parameters and so simplifying the need to
change test and inspection procedures as lead-free processes
are adopted.
Results
The key requirements of x-ray inspection of PCBs, lead-free
or otherwise, is that opens, bridges, shorts, voids, etc. can be
seen such that process and production quality control can be
Changing to a lead-free process requires a consideration of
the implications that these new materials have on PCB
manufacture and test / inspection because of their differing
properties to what has been the convention for tin/lead
solders for so many years. The lead-free learning curve has
to be climbed requiring trials to be made and errors will
occur. However, as far as x-ray inspection is concerned then
existing and future x-ray equipment will be able to see these
faults under lead-free BGA and CSPs as well as within the
rest of the SMT components. All that may be necessary is
the adjustment of certain x-ray tube parameters to ensure the
best image contrast is available to allow for easy analysis.
References
(1) www.lead-free.org; www.smartgroup.org;
www.leadfreesoldering.com
(2) Report MATC (A) 141: Code of Practice for the Use of
Electronic Components and PCBs in Lead-Free
Processing, by M. Wickham, A. Brewin, L. Zou & C.P.
Hunt, United Kingdom National Physical Laboratory
available from www.npl.co.uk
(3) X-ray tube selection criteria for BGA / CSP X-ray
inspection, D. Bernard, published in The Proceedings of
SMTA International Conference, Chicago, September
2002
(4) Selection Criteria for X-ray Inspection Systems for BGA
and CSP Solder Joint Analysis, D. Bernard, published in
The Proceedings of Nepcon Shanghai 2003
(5) www.ndt−ed.org/EducationResources/CommunityCollege/
Radiography/Physics/radmatinteraction.htm
(6) For the results of the SMART Group Lead-Free Hands On
Experience held during Nepcon UK, June 2003 see
www.smartgroup.org/ev_leadfree3.asp
Image 1: Oblique x-ray view of lead-free BGA with leadfree solder (SAC) after convection reflow, on a silver-finish
board. The balls shown with a ‘tear-drop’ shape are caused
by the deliberate inclusion of a reflow indicator. In this way,
good reflow characteristics are clearly indicated by the
unique ball shape and allow easy identification of good
process validation under x-ray inspection.
Image 2: Oblique x-ray view of lead-free BGA with leadfree solder (SAC) after convection reflow, on an OSP finish
board. The variable solder ball shape compared to image 1
suggests the process used needs to be reviewed.
Image 3: Oblique x-ray view of lead-free BGA with leadfree solder (SAC)after vapour phase reflow, on a gold-finish
board. The bridge between two solder balls is clearly visible
indicating the need to rework this part.
Image 4: Voiding within solder balls of a lead-free BGA
using lead-free solder. This level of voiding is excessive and
suggests that the reflow temperature profile used should be
modified.
Image 5: Oblique x-ray view of lead-free flip chip. The
variation in the joint shapes clearly differentiates between
open (circular) and reflowed joints (‘hourglass’ shaped).
These solder balls are ~ 190 microns in diameter.
Image 6: X-ray inspection remains valid for all SMT
components and not just BGA and CSP devices as shown in
this and images 7 and 8. In this image of a lead-free board,
there is voiding under the component that may potentially
cause problems as well as voiding in the BGA solder balls.
The insides of components are also visible under x-ray
inspection.
Image 7: Oblique x-ray view of lead-free, pin-in-hole
reflow using lead-free (SAC) solder on an OSP-finish board.
The barrels are completely filled with solder but there is
evidence of voiding within the solder.
Image 8: Oblique x-ray view of lead-free, pin-in-hole
reflow using lead-free (SAC) solder on a lead-free HASLfinish board (tin/copper/nickel finish). The barrels are not
completely filled with solder and therefore may require
rework.